ABD SOCIETY & UNC CHARLOTTE’S
DATA ANALYTICS SPEAKER SERIES
[WEBINAR]
EMBEDDING A 2,500-YEAR-OLD IDEA
THAT MARRIED DEEP LEARNING AND DOMINATED THE WORLD
Thurs, NOVEMBER 5th, 2020
12pm - 1pm
MEET our speaker
Nikolaos Vasiloglou | LinkedIn
Vice President of Data Science at Relational AI
Dr. Vasiloglou holds a PhD from the Department of Electrical and Computer Engineering at Georgia Institute of Technology. His thesis was focused on scalable machine learning over massive datasets. After graduating from Georgia Tech he founded Analytics1305 LLC and Ismion Inc. He has architected and developed the PaperBoat machine learning library which has been successfully integrated and used in the LogicBlox and HPCCSystems platforms.
In the past, he organized and Chaired MLConf and eXecutiveML, a series of conferences in machine learning held in New York, San Francisco, Seattle and Atlanta and in his spare time, continues to offer training in the U.S. through "MLTrain" and the "Learning Machine Learning by example" Meetup to integrate ML concepts in practice. Dr. Vasiloglou has served as a ML consultant to Predictix, Revolution Analytics, Damballa, Tapad and LexisNexis and he helps local Atlanta startups with machine learning systems. His work has resulted in patents and production systems
PRESENTATION TOPIC
Embedding a 2,500-Year-Old Idea That Married Deep Learning and Dominated The World
In this talk, Dr. Nikolaos will dive into the geometrical roots and deep learning. We will see how the idea of embeddings has been shaped through the centuries and what was the Eureka! moment that made deep learning the new standard of AI. He will also explain why traditional Euclidean embeddings are not sufficient and why we need to explore other curved geometries. There is an astonishing connection with what physicists use in the theory of general relativity.
WATCH RECORDING